Network Sites Working Group Meeting Report
R.Arimoto and W.Keene
Overhead 1 - Summary of progress to date
Overhead 2 - Possible network site locations in the region (see below)
Overhead 3 - Table of Japanese sites, many of which already include supporting measurements
Overhead 4 - Table of APARE sites
There was a general consensus that Cheju-do, Korea would be one of the super sites.
We then discussed the merits of installing a second super site near a dust-source region.
The site at Shapuotou was considered, but it is located near a large population and industrial center and it is impacted by local emissions. However this site is located in a major dust deposition area and, thus, could be a good location to investigate certain deposition-related processes and effects.
An alternative site near Yulin in northern Shaanxi Province, close to the border with Inner Mongolia, was also considered. B. Huebert recently visited this site. Although Yulin is a large city, air flows off the desert during spring so local impacts are probably small. A remnant of the great wall (about 30 m high) could support some limited sampling activities, but more intensive activities probably would require alternate siting to prevent damage to the ruin. Investigators from the Desert Research Institute are doing some work there apparently in collaboration with the Chinese National Academy of Sciences. Daily flights from Xi’an provide good access. The roads are also good. Power availability is uncertain but could probably be put in.
The need for multiple sampling sites in dust source regions was discussed because dust originating from different geographic area varies chemically and mineralogically. Ultimately, however, logistics and available funding will limit options.
J. Anderson and I. Solkolik emphasized the need for mineralogical in addition to chemical characterization because, for example, dust and fly ash can have identical chemical compositions.
Complications in exporting samples for analyses outside the countries in which they were collected were discussed. Export protocols are currently under discussion but all details have not yet been resolved. W. Keene mentioned one mechanism that proved workable for a previous joint Chinese-US study. Samples were split at the field site in China and analyzed independently at both Chinese and US laboratories. This protocol provided ongoing QA, fostered collegial relationships through interactions in analyzing data and resolving inconsistencies, and facilitated infrastructure development through training and technology transfer involving new techniques and procedures.
The merits of analyzing 'standard' dust material (available from the Japan) was discussed as a means of intercalibrating participating laboratories. A flier describing this reference material was distributed by M. Uematsu to interested parties.
J. Anderson argued that grab samples of surface dirt/dust in source regions are not a reliable references for the chemical and physical composition of atmospheric dust in source regions. Dust must be directly sampled in source regions to provide characterize signatures.
B. Huebert suggested we consider not putting a major portion of the US NSF funds into the Cheju-do site because substantial infrastructure already exists and the Korean programs there are well established. He recommended we consider putting the resources into other sites nearer the dust source regions.
The Chinese will support measurements near Beijing and at Qingdao.
AM. Schmoltner recommended putting a second super site near the aircraft ops center within the outflow regions. From her perspective, a great deal of effort in the form of a US supported supersite would not be needed to characterize emissions in central China.
Three potential Japanese sites were discussed:
- Oki, which has a long ~10 year data record of aerosol and trace gas composition.
- Hachijo, an island off the south east coast of Japan
Hachijo was considered the best prospect because it was farther off shore and thereby would offered a distinctly different chemical environment from Cheju-do. The island has an airport with regular flights and logistics are good. There have been some sort-term measurements of aerosol (composition?) at this location.
Haha-jima (Ogasawara Islands) was also mentioned as a possible site. More measurements have been made there but it is farther south and logistics are not as good.
Arimoto pointed out that if sites are designated as super sites it might be easier to secure resources from host countries and others to support work there.
Initially, we agreed to identify 5 supersites:
- Near Yulin in a dust source region
- Near Beijing in a heavy polluted region
- Near Qingdao at the coast
- At the Cheju-do Kosan site in a near-shore outflow region
- On Hachijo in an outflow region farther off shore.
At each of these sites, enhanced measurements including sized aerosol composition, radiation, etc. would be implemented to the extent possible with funds from national agencies.
G. Carmichael recommended that Lin’an be added to the list. It is a China MAP site located further to the south, has a long period of operation, and has a good data record. There was general agreement to add Lin’an and expand the list of supersites from 5 to 6.
There are 4 GAW sites in China
- Beijing (operated since 1983)
- Linan (operated since 1984)
- 2 other sites (one to the north near Korea and one to the west)
The Hefei site near Shanghai was discussed. Many radiation measurements are made there. The site supports various sun photometers and a micropulse lidar.
The pros and cons of various sites were discussed.
Prof. Wang said that the Yuling site would be difficult logistically.
The need to resolve data archiving was mentioned. The need to access historical data was also raised.
The optimal size cut of the 'basic' aerosol sampler was discussed at some length, mainly because the reviews of the Network SIP recommended that a 1 um ambient diameter cut would be better than a 2.5 um cut from the perspective of both understanding chemistry and radiative effects. The major points of discussion were:
- a considerable number of 2.5 um samplers may be available either gratis or at a minimal cost from IMPROVE. R. Arimoto and T. Cahill will follow-up on this.
- The 2.5 um cut was originally established by EPA as a useful benchmark based on health studies. Because many such samplers have been deployed worldwide, there is a very large data base corresponding to a 2.5 cut. In terms of atmospheric chemistry and physics, however, many feel that a 1 um cut provides better discrimination between different classes of aerosols and their associated radiative effects.
- Any specified size cut is somewhat arbitrary.
- The size cut is based on aerodynamic not optical size
- The mass median aerodynamic diameter of dust changes from about 2.5 um to about 1 um during transport. A 1 um cut would not resolve this evolution.
- The rationale for recommending a 1 um cut is based on relating dust mass to optical properties. However, such analyses require more detailed size information. Consequently, reliable analyses of this nature will be restricted to supersites where cascade impactors will be operated in parallel with measurements of aerosol optical properties and atmospheric radiation. Such coupled analyses will more than likely be restricted to intensives.
- A 2.5 cut would provide more spatial information because there are already samplers with this 'standard' cut in the region.
- Data from a 'basic' sampler configured with either a 1 or a 2.5 um cut cannot be directly related to radiative transfer. Data from such samplers are needed primarily to characterize spatial variability in chemistry.
- The group consensus was that the IMPROVE samplers with the 2.5 um cuts would provide valuable information.
Overhead - listed suite of potential aerosol and related measurements
Three classes would be measured in samples from the basic sampler: 1) Mass, 2) major ions, and 3) black carbon. Others would be restricted to intensives.
There was a brief discussion of the EC/OC issue as a result of one reviewer’s
comment that the thermal technique was unscientific. R. Arimoto cited Seinfeld’s
book indicating the method should be tied to the intended purpose of the
measurement. H. Cachier pointed out that there is a lot of variability
in measurements made with more sophisticated instruments. Intercomparisons
of methods are in progress in projects outside of ACE-Asia. J. Seinfeld
re-emphasized the need for obtaining EC/OC data for ACE-Asia.
Network Summary Report by R. Arimoto
The following draft summarizes major points mentioned by R. Arimoto during his wrap-up presentation on 14 November 1999.
We must link the facilities at the supersites with the needs of the PIs who plan to work there. This will be accomplished through the PMEL web page which will be linked to JOSS.
Arimoto will post two templates, the first describing the sites and their facilities, and the second requesting specific information from each PI including:
- A brief description of the planned activity and location.
- Power, space, etc. requirements
In addition, by early December, each PI who plans to submit a proposal as part of ACE-Asia needs to provide the following information, and this information also should be sent to the appropriate platform coordinator:
- Specific measurements, frequencies, and durations
- Other resources (e.g., towers)
- A 1-2 paragraph proposal synopsis
- Start dates
It is important to assemble the above information very soon, in part, so that we can determine when site observations will start.
The need to reliably measure wet-deposition fluxes was reemphasized. B. Huebert had mentioned previously that the lack of such measurements was a major omission in previous ACE experiments. G. Carmichael also argued strongly during his presentation that these data will be essential for constraining regional models.
A tower group was identified to address the need for towers at the two marine supersites and potential mechanisms for providing these facilities. The group includes: W. Keene (lead), R. Arimoto, M. Uematsu, and Y. J. Kim.